Generally, the accuracy of connectivity information in a distribution grid deteriorates over time after repairs, maintenance, and rebalancing efforts. Energy distributors are continually seeking automated solutions to infer the connectivity model of their distribution grids—the underlying topology between different entities such as customers, transformers, and feeders. As part of smart grid transformations, distributors are deploying different types of meters—feeder, transformer, and consumer meters. One of the expectations from this deployment is that no more field inspections would be needed to determine the connectivity model between metered entities in the future after any maintenance, such that the connectivity model would be inferred automatically. However, conventional solutions to this and related questions have proven to be inadequate.